Docosahexaenoic acid (DHA, 22:6N-3) is a major component of neural and retinal membranes. Limited accumulation is associated with learning and visual acuity deficits. DHA is transmitted from mother to the fetus/infant in the last intrauterine trimester and in milk. Premature infants may receive neither form of maternal DHA, have poor DHA stores and limited DHA synthesis from linolenic acid. Nonphysiological declines in RBC DHA occur following early delivery unless DHA is fed (human milk, fish oil). Lipid supplementation studies carried out in the Newborn Center at the University of Tennessee, Memphis will be directed toward assessing the role of specific n-6 and n-3 containing lipids in membrane biochemistry, visual acuity, recognition memory, granulocyte response to stimuli and airway function. In 1987, 195 infants in this birthweight category were discharged live from the unit. A nutrition team assures consistent unit management. 1) Preterm infants at low risk for poor developmental outcomes will be randomized to receive formula with and without DHA when >110 kcal/kg/d from formula is tolerated. Biochemical and functional followup will occur to 92 wks with assessment of plasma and red blood cell phospholipid fatty acids,k tocopherol and retinol by chromatographic techniques, retinol- binding protein by radialimmunodiffusion at enrollment, 40, 48, 57, 69, 79 and 92 wks postconception; visual acuity (Teller acuity cards) at 40, 48, 57, 69, 79 and 92 wks; recognition memory (Fagan infantest for cognitive function) at 69, 79, and 92 wks, and Bayley Mental and Physical Development (MDI/PDI) at 92 wks. Stepwise regression will include biochemical (n-3 fatty acids, retinol, -tocopherol, retinol-binding protein) and maternal/ neonatal/perinatal variables such as birth wt, gestational age, age at enrollment, O2, ventilatory support, maternal age and gravida. Ophthalmic screening will identify infants with myopia, hyperopia, astigmatism, aniosometria or strabismus precluding binocular testing. 2) Soybean phosphatidylcholine and reduced ratios of eicosapentaenoate (EPA, 20:5n-3) to DHA will be fed and their effects on PMN leukocyte composition and function studied. 3) The effect of n-3 supplementation immediately after birth on respiratory function/biochemistry will be studied in very small premature infants (<1000 g birth wt) who are at greatest risk for chronic lung disease. Their n-3 status is very poor. Evidence suggests that polyunsaturated fatty acids, especially fish oil DNA and EPA, could protect against lung damage induced by artificial ventilation and oxygen administration. DHA administration in intravenous lipid and formula will begin within 48-72 hrs after birth. In 1987, 77 of 139 inborn infants (55%^) weighing less than 1000 g at birth were discharged live from the UT, Memphis, Newborn Center.